What is my objective giving this formation

• Show basic usage of Rmarkdown
• Share some personnal advices and reflexions about its usage

Why Rmarkdown?

In my M.Sc. Analysis folder I have… 313 documnents with a “.R” extension!

How many time do you think I searched in more than five folder to find the code to reproduce a result? How many times I open the correct R file without finding the code of interest?

How many duplication do you think I have in this mess… It’s about Git, but it is another story.

Rmarkdown allow you to :

• Clearly write your way of thinking when trying something/analysing your data set
• Display reproducible code for anyone interested about your data/results/analysis (Research Director??)
• Produce professionnal level documents with figures without worrying about formating (Supplementary material!!)
• Create interactive documents!

In one sentend, Rmwarkdown favor reproducibility, communication, structured thinking, and aesthetism.

Rmarkdown has some advantages :

• It is easy (“If you can write an emoticon, you can write RMarkdown”)
• It is quick (far more than producing plots at the right dimensions, then copy-pasting to a word document, then screaming on your computer about layout)
• Allow you to write automatically internet-friendly document (even books!), without having to worry about html, css and so on.

• It produces full plain-tet files which works well with version control! I see one con :

• We (partially) loose the reviewing facility of word.

Let’s begin!

Chunks

Code can be displayed inside chunks. They are delimited by ```{r} and ```.We can insert chunk in the following ways :

• Ctrl + Alt + i (or Cmd + Alt + i)
• Add chunk command in the editor bar
• Code > Insert chunk

Let’s try something very simple

plot(iris)

ggplot(iris, aes(x = Sepal.Length, y = Sepal.Width, color = Species)) +
  geom_point() +
  theme_minimal()

cor( c( 1 , 1 ), c( 2 , 3 ) )

## [1] NA

Other languages

It is interesting to know that chunks can highlight and run code of other languages, like

• Bash

hostname

## lucasd-HP-EliteBook-2560p

• Python

x = "Hello, Python!"
print(x.split(" "))

## ['Hello,', 'Python!']

• stan

// saved as 8schools.stan
data {
  int<lower=0> J;         // number of schools
  real y[J];              // estimated treatment effects
  real<lower=0> sigma[J]; // standard error of effect estimates
}
parameters {
  real mu;                // population treatment effect
  real<lower=0> tau;      // standard deviation in treatment effects
  vector[J] eta;          // unscaled deviation from mu by school
}
transformed parameters {
  vector[J] theta = mu + tau * eta;        // school treatment effects
}
model {
  target += normal_lpdf(eta | 0, 1);       // prior log-density
  target += normal_lpdf(y | theta, sigma); // log-likelihood
}

Inline code

It is possible to display the result of a code in directly in the text. We will exemplify that with an lm exemple

fit <- lm(data = iris, Sepal.Length ~ Sepal.Width)
coef(fit)

## (Intercept) Sepal.Width
##   6.5262226  -0.2233611

Globally, without accounting for among species differences, a 10 cm increase of Iris’s sepal width imply a 0.22 cm decrease of sepal length.

Classical kable

Publication quality table

One can format very easily publication level table in pdf document using latex. In html, we have to use a little trick and use ghostscipt and kableExtra.

Exercises

• Could you display the relevant information to understand and evaluate the results of the above linear regression?
• stargazer package produces automatically interesting tables to compare models, would you try it with the best and second best models?

Documents

• html_notebook - Interactive R Notebooks
• html_document - HTML document w/ Bootstrap CSS
• pdf_document - PDF document (via LaTeX template)
• word_document - Microsoft Word document (docx)
• odt_document - OpenDocument Text document
• rtf_document - Rich Text Format document
• md_document - Markdown document (various flavors)

Presentations

• ioslides_presentation - HTML presentation with ioslides
• revealjs::revealjs_presentation - HTML presentation with reveal.js
• slidy_presentation - HTML presentation with W3C Slidy
• beamer_presentation - PDF presentation with LaTeX Beamer
• powerpoint_presentation - PowerPoint presentation

An exemple of presentation can be downloaded here.

Table of content

One can include pretty beatifull table of content by modifying the yaml options!

Exercise

• Would you try to transform the present document into a decent presentation?

Inline equations

One can insert inline latex equations, just by using $. For example, \(y_i = \beta_1 x_i + \beta_0 + \epsilon_i\) represents a simple linear equations. Thus wonderfull functionality allow to use any mathematical notation. This is useful to indicate that \(\beta_1\) is the slope and \(\beta_0\) the intercept!

Block equations

One can insert block latex equations using $$. For example, one can estimate the parameters of a linear model containing \(p\) predictors (\(p < n\)) using the following linear algebra formulation and a little bit of linear algebra.

\[ \mathbf{Y} \begin{bmatrix} Y_1 \\ \vdots \\ Y_n \end{bmatrix} = \mathbf{X} \begin{bmatrix} 1 & x_{1,2} & \cdots & x_{1,p} \\ 1 & \vdots & \ddots & \vdots \\ 1 & x_{n,2} & \cdots & x_{n,p} \\ \end{bmatrix} \mathbf{\beta} \begin{bmatrix} \beta_0\\ \beta_1 \end{bmatrix} + \mathbf{\epsilon} \begin{bmatrix} \epsilon_1\\ \vdots\\ \epsilon_n\\ \end{bmatrix} \]